Computing Device Topology-Aware Availability Modeling And Computing Resource Provisioning

Abstract

Methods, systems, and apparatus, including computer-readable storage media and computer program products for determining thresholds for guaranteeing the availability of partitions of different quantities of computing devices and using those thresholds to more accurately provision computing resources to users of a computing platform. A partition (also called a “slice”) of computing devices can be provisioned to a user of a computing platform. Slice-size thresholds can indicate how many computing devices on the platform can be provisioned as part of a single slice. Determining slice-size thresholds before provisioning computing resources reduces or eliminates the chance that requests to provision a slice of computing devices are approved when the platform is not able to fulfill those requests with the largest slice-size possible for a given request.

1 . A method for provisioning partitions of groups of computing devices, the method comprising:
 determining, by one or more processors, one or more combinations of computing devices from each group of a plurality of groups of computing devices, at least one combination of the one or more combinations comprising quantities of computing devices from the plurality of computing devices;   receiving, from the one or more processors, a first partition size;   identifying, by the one or more processors and from the one or more combinations, a threshold value corresponding to a quantity of partitions of the first partition size that can be partitioned from a combination of the one or more combinations; and   provisioning, by the one or more processors, a partition of computing devices of a size equal to the first partition size, the provisioning based at least on the threshold value.   
     
     
         2 . The method of  claim 1 , wherein a group of computing devices is communicatively coupled through an inter-chip-interconnect. 
     
     
         3 . The method of  claim 2 , further comprising provisioning partitions of computing devices corresponding to the largest partition size of computing devices that meets or exceeds requested quantities of computing resources received by the one or more processors and does not cause a total quantity of provisioned partitions across the plurality of groups of computing devices to exceed the threshold value. 
     
     
         4 . The method of  claim 1 , wherein:
 a quantity of computing devices for a group represented in the combination is greater than a respective threshold number of computing devices in the group meeting a predetermined uptime threshold, and   a sum of the quantities of computing devices for the plurality of groups represented in the combination is approximately equal to a minimum number of computing devices of the plurality of groups meeting the predetermined uptime threshold.   
     
     
         5 . The method of  claim 4 , wherein the predetermined uptime threshold is a percentage of uptime for a computing device over a specified period of time. 
     
     
         6 . The method of  claim 1 , wherein:
 the plurality of groups comprises one or more full groups and one or more partially full groups;   a sum of the quantities of computing devices for the plurality of groups represented in the combination is a sum of quantities of computing devices represented in the combination for the one or more full groups plus a sum of average quantities of computing devices in the one or more partially full groups minus a respective difference between the a size of a full group and the size of each of the one or more partially full groups meeting a predetermined uptime threshold; and   the sum of the quantities is approximately equal to a minimum number of computing devices of the plurality of groups meeting the predetermined uptime threshold.   
     
     
         7 . The method of  claim 6 , further comprising updating, by the one or more processors, a current quantity of partitions that are provisioned from the computing devices and of the first partition size. 
     
     
         8 . The method of  claim 1 , wherein the threshold value corresponds to a smallest quantity of partitions of the first partition size that can be partitioned from any combination of the one or more combinations. 
     
     
         9 . The method of  claim 8 , further comprising:
 receiving, by the one or more processors, a request to provision the partition of the first partition size;   determining, by the one or more processors, that the threshold value has not been exceeded; and   provisioning, by the one or more processors, the partition of the first partition size.   
     
     
         10 . The method of  claim 1 , wherein a computing device of the computing devices comprises a plurality of computing devices interconnected according to a network topology. 
     
     
         11 . A system for provisioning partitions of groups of computing devices, the system comprising one or more processors configured to:
 determine one or more combinations of computing devices from each group of a plurality of groups of computing devices, at least one combination of the one or more combinations comprising quantities of computing devices from the plurality of computing devices;   receive a first partition size;   identify, from the one or more combinations, a threshold value corresponding to a quantity of partitions of the first partition size that can be partitioned from a combination of the one or more combinations; and   provision a partition of computing devices of a size equal to the first partition size, the provisioning based at least on the threshold value.   
     
     
         12 . The system of  claim 11 , wherein a group of computing devices is communicatively coupled through an inter-chip-interconnect. 
     
     
         13 . The system of  claim 12 , wherein the one or more processors are further configured to provision partitions of computing devices corresponding to the largest partition size of computing devices that meets or exceeds requested quantities of computing resources received by the one or more processors and does not cause a total quantity of provisioned partitions across the plurality of groups of computing devices to exceed the threshold value. 
     
     
         14 . The system of  claim 11 , wherein:
 a quantity of computing devices for a group represented in the combination is greater than a respective threshold number of computing devices in the group meeting a predetermined uptime threshold, and   a sum of the quantities of computing devices for the plurality of groups represented in the combination is approximately equal to a minimum number of computing devices of the plurality of groups meeting the predetermined uptime threshold.   
     
     
         15 . The system of  claim 14 , wherein the predetermined uptime threshold is a percentage of uptime for a computing device over a specified period of time. 
     
     
         16 . The system of  claim 11 , wherein:
 the plurality of groups comprises one or more full groups and one or more partially full groups;   a sum of the quantities of computing devices for the plurality of groups represented in the combination is a sum of quantities of computing devices represented in the combination for the one or more full groups plus a sum of average quantities of computing devices in the one or more partially full groups minus a respective difference between the a size of a full group and the size of each of the one or more partially full group meeting a predetermined uptime threshold; and   the sum of the quantities is approximately equal to a minimum number of computing devices of the plurality of groups meeting the predetermined uptime threshold.   
     
     
         17 . The system of  claim 16 , wherein the one or more processors are further configured to update, by the one or more processors, a current quantity of partitions that are provisioned from the computing devices and of the first partition size. 
     
     
         18 . The system of  claim 11 , wherein the threshold value corresponds to a smallest quantity of partitions of the first partition size that can be partitioned from any combination of the one or more combinations. 
     
     
         19 . The system of  claim 18 , wherein the one or more processors are further configured to:
 receive a request to provision the partition of the first partition size;   determine that the threshold value has not been exceeded; and   provision the partition of the first partition size.   
     
     
         20 . One or more non-transitory computer-readable storage media, storing instructions that when executed by one or more processors, causes the one or more processors to perform operations comprising:
 determining one or more combinations of computing devices from each group of a plurality of groups of computing devices, at least one combination of the one or more combinations comprising quantities of computing devices from the plurality of computing devices;   receiving a first partition size;   identifying, from the one or more combinations, a threshold value corresponding to a quantity of partitions of the first partition size that can be partitioned from a combination of the one or more combinations; and   provisioning a partition of computing devices of a size equal to the first partition size, the provisioning based at least on the threshold value.